학술논문
Evaluation of the NORRISK2 model for cardiovascular risk prediction in South-Asians living in Norway.
Document Type
Article
Author
Source
Subject
*CARDIOVASCULAR diseases risk factors
*HOSPITAL admission & discharge
*CARDIOVASCULAR disease diagnosis
*CAUSES of death
CARDIOVASCULAR disease related mortality
*
*
*
Language
ISSN
0803-2491
Abstract
Introduction: The NORRISK 2 model estimates an individual's 10-year risk of cardiovascular disease (CVD) based on known risk factors. It was developed from the Cohort of Norway (CONOR) surveys 1994-1999 linked to CVDNOR, a database of cardiovascular hospital discharge diagnoses and mortality in Norway in 1994-2014. The model is included in Norwegian guidelines, but has not been validated in immigrants in Norway. Aims: To evaluate the NORRISK 2 model in immigrants from South Asia and compare with Norwegian born men and women. Methods: We included participants aged 30-74 years from The Oslo Health Study (HUBRO), The Oslo Immigrant Health Study (IHUBRO) and the Romsås in Motion Study (MoRo II), all part of CONOR. The study population consists of 7823 women and 6429 men born in Norway and 832 women and 1192 men born in South Asia, linked to CVDNOR and the Norwegian Patient Registry. An endpoint was defined as the first occurrence of hospitalization with AMI or acute cerebral stroke or death from coronary heart disease or death from acute cerebral stroke as underlying cause. Due to limited number of cases, we calculated13-year risk of CVD instead of the original 10-year risk by the NORRISK 2 model. We validated the model by Harrell's C and calibration plots based on predicted minus observed cumulative risk within quintiles of predicted risk. Results: The predicted 13-year risk by the NORRISK 2 model was 4.0% (95% CI 3.7-4.3) compared with observed 7.2% (5.8-9.0) in South Asian men and 1.1% (1.0-1.2) predicted versus 2.6% (1.6-4.0) observed in South Asian women. Corresponding results for Norwegian-born were 6.4% (6.2-6.6) predicted versus 5.9% (5.3-6.5) observed in men and 2.8% (2.7-2.8) predicted versus 2.3% (2.0-2.7) observed in women. NORRISK2 underestimated risk in all quintiles of predicted risk in South Asian men and women. Harrell's C was 0.79 in men and women from South-Asia and 0.79 in Norwegianborn men and 0.81 in Norwegian-born women. Conclusions: NORRISK2 underestimated 13-year CVD risk in South-Asians. Discrimination ability was good. An updated model for men and women born in South Asia is warranted. [ABSTRACT FROM AUTHOR]